Part I — Situation overview

On Friday, 25 September 2026, the Hungarian Central Statistical Office (KSH) published the August vital statistics and the labour-market flash report on the same day. In August 5,922 children were born, 7.4 per cent fewer than a year earlier — for the first time in the history of the data series, the number of births in an August month did not even reach six thousand. 9,452 people died, so the natural decrease, i.e. the excess of deaths over births, was 3,530, compared with 2,968 in August last year. The number of marriages fell by 12 per cent. The total fertility rate — the estimated number of children a woman would have over her lifetime at the given year’s frequencies — was 1.29 in the first eight months of the year, compared with 1.31 a year earlier. On the same day, according to the KSH, 4 million 654 thousand people were in work in August and 229 thousand were unemployed. The unemployment rate was 4.7 per cent, 0.3 percentage points (pp) higher than a year earlier.

Taken separately, neither data series is a new topic for MIAK. MIAK wrote about the targeting of family support in April (Otthon Start, mothers’ PIT exemption, family tax allowance) and in mid-September (Family allowance unchanged since 2008). It published analyses on data-based management of the labour market in June (Guest-worker stop) and on the burdens of the pension system in August (A minimum pension of HUF 120 thousand). What is new now is that the two processes point in the same direction. According to Portfolio, between 2010 and 2020 an average of 7,500 children were born per month; today it is barely more than 5,500. ING Bank’s chief economist, Péter Virovácz, told ATV that the supply side of the labour market has shrunk by 180 thousand people since mid-2022. The three-month moving average of unemployment — the average of three consecutive months, which smooths out monthly fluctuation — has nevertheless risen to 4.8 per cent, only 0.1 pp below the ten-year peak of late 2025. Every third unemployed person has been looking for a job for more than a year. According to Gábor Regős, senior economist at Gránit Asset Management, in one year the working-age population fell by 60 thousand and the number of economically active people (those in work and those actively seeking a job) by 8 thousand, while the number of unemployed rose by 8 thousand.

MIAK’s reading: fewer people, yet more unemployed — this is not merely a fall in demand, but a sign that the economy is unable to retain the existing workforce and channel it to the right place. The demographic side acts slowly but irreversibly: the current monthly number of births means cohorts roughly a quarter smaller than the 2010s average in the labour market of the 2040s. The labour supply of the next two decades will therefore be decided less by the number of births than by how many people the economy keeps active and how many it allows to drop out permanently. In a shrinking country every long-term unemployed person costs money twice: they are missing from production, and they are missing from among the contributors to the pension system.

Part II — Foundations in the literature

Three sources provide the interpretative framework. According to the European Commission’s 2026 macroeconomic report (2026 Macroeconomic Report), slow productivity growth and unfavourable demography together hold back growth, and also threaten the financial sustainability of welfare systems. The report’s answer is to activate groups under-represented in the labour market, supplemented by bringing in foreign workers. The Commission’s thematic analysis of pensions (Adequacy and Sustainability of Pensions) directly links the demographic and labour-market sides. According to it, because of low fertility and longer life expectancy, large working-age cohorts are being replaced by smaller ones. The adequacy of pensions therefore depends increasingly on long, uninterrupted careers, so reducing long-term youth unemployment is itself pension policy. Martin Ford (an American author who spent more than twenty-five years in software development) describes the self-reinforcing nature of long-term unemployment in his book Rise of the Robots (Hungarian edition: Robotok kora). Professional skills erode, employers screen out those who have been jobless for longer, and after roughly six months the prospects deteriorate drastically. The detailed discussion of the literature — by author, with quotations — can be found in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. The first makes reading the two data series together a regular practice, the second slows permanent exit from the labour market, and the third extends the measurement of family support to include its effect on parents’ labour-market participation.

3.1 Quarterly combined population and labour balance (first edition by 31 December 2026, quarterly thereafter)

MIAK proposes that the KSH and the Ministry of Social Affairs and Families, which is responsible for employment policy, publish a joint public balance every quarter on why labour supply — i.e. the combined number of those in work and those actively seeking a job — is changing. The balance should break down the change into four components:

  • the decline of the working-age population (demographic effect);
  • the change in participation (how many people enter or leave the labour market);
  • international migration (including workers arriving from third countries);
  • the development of unemployment and permanent exit.

The methodology should be developed by the KSH, with respect for its professional independence. The Government’s task is to provide the funding and to establish the legal basis for the data links (tax authority, pension insurance and employment registers). The balance should show, in one place and with explanations, the raw monthly figure, the three-month average and the seasonally adjusted figure — the version from which the usual fluctuation tied to the calendar period has been filtered out. This week the press simultaneously reported that the number of people in employment had grown and that it had fallen by 21 thousand from July to August. The discrepancy presumably stems from different indicators, but in public debate it gets blurred. The balance should be a joint module of the DM1 demographic and the FO1 employment data platform, broken down by county. According to the European Commission’s report (see 6.4.1), demography and productivity together determine growth. Today the Hungarian public cannot see this interaction in a single table.

3.2 A six-month activation threshold against long-term unemployment (launch on 1 January 2027, full coverage by 30 June 2027)

MIAK proposes that every registered jobseeker who has been out of work for six months receive a tailored offer from the public employment service within 30 days. The offer can take three forms: retraining adapted to local demand under the FO3 targeted retraining programme, a fixed-term wage subsidy to the employer, or relocation and commuting support. The six-month threshold is not arbitrary. According to the American experiment described by Ford (see 6.4.3), beyond this threshold employers invite even a candidate with relevant experience to interview less often than an inexperienced applicant who has only just become jobless. In Hungary today every third unemployed person has been looking for work for more than a year, so the intervention is already arriving late. The second element of the proposal is a rapid response to layoffs. Under the Labour Code, employers are already obliged today to inform the public employment body of collective redundancies. Within 15 days of such a notification, on-site counselling and a retraining offer should begin, even before the dismissal takes effect. The third element is the activation rule: if the three-month average of unemployment exceeds 5.0 per cent in two consecutive months, the FO7 anticyclical employment fund should be triggered. According to the EU analysis of pensions (see 6.4.2), early and lasting labour-market integration is also a condition for the sustainability of the pension system.

3.3 A dual-criteria impact assessment of family support (methodology by 31 December 2026, first report by 31 March 2027)

The DM2 family-support impact assessment has so far primarily asked how many additional births a form of support brings. MIAK proposes that for every element the assessment also answer a second question: how it affects the labour-market participation and return of parents, especially mothers. In a country with a shrinking workforce, a form of support that brings births but permanently withdraws the parent from the labour market partly works against itself. A form of support, however, that eases both having children and returning to work pays off twice. The assessment should cover mothers’ personal income tax (PIT) exemption, childcare benefits, the baby-expecting loan (babaváró) and the availability of nursery places. The latter is one of the pillars of the FO6 demographic response strategy. The report’s results should also be built into the SZ13 pension adequacy and sustainability calculation. The long-term path of the pension system must be recalculated with the fertility rate of 1.29 and with labour supply that has shrunk by 180 thousand people since 2022, not with the earlier, more favourable assumptions. In line with the G1 data-driven budgeting principle, the dual-criteria measurement should also be linked to the family-policy lines of the 2027 budget.

The common principle of the three proposals is that in a shrinking country the scarcest resource is people who are able and willing to work. Policy can influence the number of births only slowly and uncertainly. What can be achieved in the coming years, however, is that those who are in the labour market today do not drop out of it permanently. In the European Commission’s words, this is activating the untapped labour reserve (see 6.4.1), and its precondition is that we see demographic and labour-market data in a single balance.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Labour market Fewer people slide into long-term unemployment; re-employment after layoffs is faster Training providers may pick the easily placeable participants, leaving out those in the most difficult situation
Demography The evaluation of family support measures parents’ return to work alongside the number of births Demonstrating the birth effect takes years; early results can be misinterpreted
Budget Activation is cheaper than years of benefits and lost contribution revenue In an economic slowdown the 5.0 per cent threshold can quickly trigger significant spending
Public administration The quarterly balance brings today’s scattered data into one place Building the data links requires data-protection safeguards, and the capacity of the employment service is limited
Public life The debate over interpreting the numbers shifts towards measurable causes A bad quarterly figure in the balance can become a political weapon, regardless of the causes

The main question to weigh is the targeting of the six-month threshold. If the offer is only formal, for example a mandatory course that does not match local demand, the system is fulfilled on paper while unemployment remains. That is why MIAK proposes that training providers’ fees be linked to job placement, with a higher fee for participants in the most difficult situation. The second question is the dual measurement of family support. The effect of a form of support on births becomes visible only years later, its labour-market effect sooner. There is a danger that the faster result overshadows the slower one, so the report must present both effects over the same time horizon. The third question is timing. According to some labour-market analysts, the rise in unemployment is partly temporary, an effect of the reshuffling after the change of government. If so, early intervention is cheaper because it affects fewer people. If not, it is precisely early intervention that prevents temporary unemployment from becoming long-term.

Part V — Measurability and summary

5.1 What is worth following? (suggested KPIs)

MIAK suggests following these performance indicators (KPIs, Key Performance Indicators):

  • Share of the long-term unemployed: the share of those who have been looking for a job for more than a year among the unemployed. According to ING’s analysis, it is currently about one third. Suggested target: below 25 per cent by the end of 2027.
  • Coverage of the activation offer: what percentage of jobseekers reaching the six-month threshold receive a tailored offer within 30 days, and how many of them are working six months later. Suggested target: at least 90 per cent coverage from mid-2027.
  • Participation adjusted for the demographic effect: the share of the economically active within the working-age population. It is worth following whether the number of active people is falling more slowly than the working-age population.
  • Publication of the combined balance: the first edition by 31 December 2026, quarterly thereafter, coordinated with the KSH flash reports.
  • Dual-criteria impact report: by 31 March 2027, published birth and labour-market impact estimates for at least four family-support elements.

5.2 Summary

MIAK’s request is that the number of births and unemployment should not be two separate news items, but two sides of a single balance. The Government and the KSH should show every quarter why the workforce is shrinking. The employment service should give every jobseeker a genuine offer after six months. And family support should also be measured by whether it keeps parents in the labour market. No single measure will reverse the decline in births in the short term. What can be achieved in the coming years, however, is that the existing workforce does not shrink faster than the population.

Two MIAK foundational values are at play here. Data-drivenness, because the two data series are published separately today, and the debate goes on without interpreting the numbers. Yet the registers needed for the decomposition exist; they are just not linked. And ideology-free decision-making, because on population and unemployment political actors look for each other’s responsibility. The balance, by contrast, shows which cause acts with what weight — and that is independent of who governs.

Part VI — Reasoning and further sources

6.1 Press framing by spectrum

In the left-liberal band, 444.hu framed the birth figure as extraordinary (“shockingly few children”), but without evaluative political commentary, confining itself to the KSH figures. It highlighted the accelerating decline in August, and also mentioned that the fertility rate had improved slightly compared with July. HVG’s subscription analysis (the article was not publicly downloadable) examined the other end of demography, the pension system. It called the HUF 120 thousand minimum pension generous, but listed numerous open questions, and linked the delayed announcements to the budget’s narrow room for manoeuvre. The link between the number of births and unemployment did not appear in either outlet.

In the general public-affairs band, 24.hu presented the eight-month natural decrease with a tangible image: it roughly equals half the population of Baja. It added that the number of deaths had fallen so far this year. ATV, by contrast, used a question headline to link the rise in unemployment to the change of government. The article quoted three analysts, of whom the Gránit Asset Management economist mentioned the post-election reshuffling as a possible fourth cause, alongside demography, the stagnating economy and labour arriving from third countries. ATV’s article was the only one that discussed the rise in unemployment in the context of the shrinking population and falling participation.

In the economic band, Portfolio presented the birth figure as “lousy” data. It emphasised that the summer months usually bring the most births, so the seasonally adjusted figure paints an even worse picture. Its analysis of the labour market (the article was not publicly downloadable) called the processes “contradictory”, because the number of employed and of unemployed rose at the same time.

In the conservative band, Magyar Nemzet chose a personal framing. It focused on Prime Minister Péter Magyar’s laughing-emoji reaction, to which the head of government added that Fidesz would shift the responsibility onto the next government. The paper also criticised the Prime Minister for not mentioning the rise in unemployment. In another article the paper quoted a Fidesz Member of Parliament, according to whom almost 1,700 jobs are being lost. The paper also pointed out that, according to the Ministry of Social Affairs and Families, a decision had already been made on the post of state secretary for employment, but the appointment had not yet been announced and the name had not been made public. From MIAK’s point of view, this framing is substantive insofar as the absence of an institutional owner of employment policy is a real risk, precisely at a time when a rapid labour-market response would be needed.

6.2 Facts and data

Data Value Source
Live births, August 2026 5,922 (−7.4% year on year) KSH, 444.hu and 24.hu, 25 September 2026
Deaths, August 2026 9,452 (+0.9%, 87 more) KSH, Portfolio, 25 September 2026
Natural decrease, August 2026 3,530 people (August 2025: 2,968 people) KSH, 444.hu, 25 September 2026
Marriages, August 2026 5,633 (−12%, 739 fewer) KSH, 24.hu, 25 September 2026
Live births and deaths per thousand inhabitants, August 7.4 and 11.8 respectively KSH, 24.hu, 25 September 2026
Live births, January–August 46,497 (−3.5%, 1,663 fewer) KSH, 24.hu, 25 September 2026
Deaths, January–August 81,397 (−2.6%) KSH, 24.hu, 25 September 2026
Natural decrease, January–August 34,900 people (same period of 2025: 35,374 people) KSH, 24.hu, 25 September 2026
Total fertility rate, January–August 1.29 (1.31 a year earlier; this year’s July value 1.28) KSH, 444.hu and 24.hu, 25 September 2026
Average monthly number of births 2010–2020: approx. 7,500; today barely more than 5,500 Portfolio, 25 September 2026
Employed (aged 15–74), August 2026 4 million 654 thousand KSH, Portfolio, 25 September 2026
Unemployed, August 2026 229 thousand KSH, Portfolio, 25 September 2026
Unemployment rate, August 2026 4.7% (+0.3 pp year on year, +0.2 pp month on month) KSH, Magyar Nemzet, 25 September 2026
Unemployment rate, three-month moving average (June–August) 4.8% (0.1 pp below the ten-year peak of late 2025) ING Bank, ATV, 26 September 2026
Change in labour supply since mid-2022 −180 thousand people ING Bank, ATV, 26 September 2026
Share of those seeking a job for more than a year every third unemployed person ING Bank, ATV, 26 September 2026
Change in one year working-age population −60 thousand, economically active −8 thousand, unemployed +8 thousand Gránit Asset Management, ATV and Magyar Nemzet, 25–26 September 2026
Expected unemployment rate for 2026 around 4.6% MBH Bank, ATV, 26 September 2026
Minimum pension from January 2027 from HUF 28,500 to HUF 120 thousand HVG, 26 September 2026

Two remarks belong to the figures. First, the sources give a differing picture of the monthly change in employment. Portfolio highlighted a slight monthly increase in the number of employed. According to the analysis quoted by ATV, however, the number of employed fell by 21 thousand from July to August: 14 thousand left the labour market and 7 thousand became unemployed. The discrepancy probably stems from the use of different indicators (raw monthly, adjusted or averaged data) — which itself argues for the combined balance proposed in point 3.1. Second, the seasonality of the birth figure can be misleading. At first glance, the August figure of 5,922 is above the monthly average of the past year, but summer traditionally brings the most births, so the 7.4 per cent annual decline is the more accurate signal.

6.3 Policy dimensions

  • Demography (programme points and background material) — public, linked demographic data and the impact assessment of family support, which MIAK now extends with the effect on labour-market participation (programme point ID: DM1, DM2);
  • Employment policy and labour market (programme points and background material) — employment data platform, targeted retraining, demographic response strategy and anticyclical employment fund; according to the background material, spending on active labour-market instruments in Hungary is below the EU average (programme point ID: FO1, FO3, FO6, FO7);
  • Social policy (programme points) — the adequacy and sustainability of the pension system with a shrinking contributor base (programme point ID: SZ13);
  • Economy (programme points) — itemised, measurable presentation of family-policy and employment spending in the budget (programme point ID: G1).

6.4 Literature in detail

6.4.1 European Commission: 2026 Macroeconomic Report

According to the European Commission’s 2026 macroeconomic report, European growth is held back by two factors together: slow productivity growth and unfavourable demography. The report compares the Union with the United States. Since 2000 the US population has grown by roughly 20 per cent and the EU’s by 5 per cent, while the labour force has grown by 18 per cent in the United States and by 12 per cent in the Union. According to the report, overall economic growth matters not only for living standards, but also for debt sustainability and for financing expenditure that does not depend on population size. Its conclusion on the demographic outlook:

“Productivity growth at the pace seen since 2015, combined with the demographic projections, would only be enough to keep the EU economy’s real GDP flat until 2050.”

In response, alongside stimulating private investment, the report urges the activation of groups under-represented in the labour market, supplemented by attracting foreign talent. It regards well-designed social protection and effective labour-market activation together as an instrument that brings the untapped labour reserve back into the labour market. Translated to the Hungarian situation: if the demographic side is even less favourable than the EU average, then bringing back those who have dropped out of the labour market is not a social-policy add-on but a precondition for growth. That is why MIAK proposes in point 3.1 that demographic and labour-market change be presented in a single balance, broken down into components.

📖 Source: European Commission: 2026 Macroeconomic Report (European Economy Institutional Paper 328)

6.4.2 European Commission: Adequacy and Sustainability of Pensions

According to the Commission’s thematic analysis prepared under the European Semester, pension policy has three objectives: adequate old-age income, financial sustainability and the highest possible employment. Population ageing will make all three harder in the coming decades. The analysis puts it this way:

“The transition from large working-age cohorts to smaller ones, resulting from low fertility and rising life expectancy, requires adjustment in retirement practices and pension systems in all Member States.”

According to the analysis, an adequate pension depends increasingly on the number of years spent contributing. The replacement rate — the ratio of the pension to previous earnings — of those with careers shorter than 40 years lags significantly behind that of those completing a full career. Attention therefore cannot be confined to the last stage of the career: reducing long-term youth unemployment and early, regular labour-market integration also mitigate pension risk. The analysis also warns that earlier early-retirement, disability and unemployment-benefit schemes often encouraged precisely early exit from the labour market. Two conclusions follow for the Hungarian case. Action against long-term unemployment (3.2) is also pension policy. And the sustainability calculation of the pension system (3.3) must be updated with the current fertility rate of 1.29.

📖 Source: European Commission: European Semester Thematic Factsheet — Adequacy and Sustainability of Pensions

6.4.3 Martin Ford: Rise of the Robots

The main theme of Ford’s book is technological unemployment, but its chapter on the labour market also describes the general mechanism of long-term unemployment. According to Ford, prolonged joblessness reinforces itself: professional skills erode over time, motivation declines, and some employers explicitly discriminate against the long-term unemployed, often without even reading their CVs. Ford describes an American field experiment conducted by Rand Ghayad, a doctoral student at Northeastern University in Boston. According to the experiment, a candidate who had only just become jobless but had no relevant experience was more likely to be invited to interview than one who had relevant experience but had not worked for more than six months. Citing a research institute report, Ford adds that the long-term unemployed do not differ substantially from other workers: their situation is largely a matter of bad luck, i.e. of whether they lost their job at a bad time. Ford also writes that in the United States the fall in the participation rate cannot be explained by demography alone, and that disability benefits have become a channel for leaving the labour market for many. The same two phenomena are visible in the Hungarian data: every third unemployed person has been looking for work for more than a year, and some of those leaving employment do not become unemployed but leave the labour market. That is why MIAK proposes in point 3.2 that the tailored offer arrive at the latest when the six-month threshold is reached, and not months afterwards.

📖 Source: Martin Ford: Robotok kora (original title: Rise of the Robots — Technology and the Threat of a Jobless Future, 2015)

6.5 International comparison

According to the European Commission’s 2026 report, the EU employment rate rose to a record 76.1 per cent in the first quarter of 2025, partly owing to the high labour-market participation of young people and women. Employment in the Union can thus grow even when the population barely grows — if a larger share of the working-age population works. The same report shows that since 2000 the labour force has grown by 18 per cent in the United States and by 12 per cent in the Union, and that the American growth advantage stems partly from this faster population and labour-force growth. Based on 2016 data, the EU analysis of pensions shows that the employment rate of older workers (aged 55–64) ranged from 36.3 per cent in Greece to 75.5 per cent in Sweden, against an EU average of 55.3 per cent. In the same year women had careers on average 4.9 years shorter than men. These differences indicate that the size of the labour reserve in each country depends primarily not on population size but on the level of participation. According to MIAK’s employment-policy background material, Denmark spends about 2 per cent of gross domestic product (GDP) on active labour-market instruments, the EU average is about 0.5 per cent, and Hungary about 0.3 per cent. The Danish system pairs easy dismissal with high unemployment benefits and intensive retraining, and achieves one of the lowest long-term unemployment rates in the Union — albeit at a high cost.

Demography

  • DM1 — Demographic data platform
  • DM2 — Impact assessment of family support

Employment policy and labour market

  • FO1 — Employment data platform
  • FO3 — Targeted retraining programme
  • FO6 — Demographic response strategy
  • FO7 — Anticyclical employment fund

Social policy

  • SZ13 — Pension adequacy and sustainability

Economy

  • G1 — Data-driven budget

Proposed new programme point: Six-month activation threshold — every jobseeker registered for six months receives a tailored retraining, wage-subsidy or mobility offer within 30 days, and after collective redundancies rapid-response counselling starts within 15 days — for the Employment policy and labour market area.

6.7 List of sources

Press sources (MIAK press monitor, 26 September 2026 — topic 1):

Knowledge base references (books):

  • 📖 European Commission: 2026 Macroeconomic Report (European Economy Institutional Paper 328)
  • 📖 European Commission: European Semester Thematic Factsheet — Adequacy and Sustainability of Pensions
  • 📖 Martin Ford: Robotok kora (Rise of the Robots, 2015)

MIAK internal materials:

  • MIAK policy area: Demography (programme points; programme point ID: DM1, DM2)
  • MIAK policy area: Demography (background material — the links between fertility, housing and the labour market)
  • MIAK policy area: Employment policy and labour market (programme points; programme point ID: FO1, FO3, FO6, FO7)
  • MIAK policy area: Employment policy and labour market (background material — international comparison of active labour-market spending)
  • MIAK policy area: Social policy (programme points; programme point ID: SZ13)
  • MIAK policy area: Economy (programme points; programme point ID: G1)
  • MIAK press monitor, 26 September 2026 — topic 1, score: 90/100

Supplementary public data sources:

  • KSH — vital statistics flash report, August 2026
  • KSH — labour-market flash report, August 2026
  • ING Bank, Gránit Asset Management, MBH Bank — labour-market analyst commentaries (as reported by ATV)

Generation metadata